LRC: Dependency-Aware Cache Management for Data Analytics Clusters

TL;DR

This paper introduces LRC, a cache management policy that leverages data dependency DAGs in data analytics clusters to improve cache efficiency and significantly speed up applications compared to traditional LRU.

Contribution

The paper presents LRC, a novel cache replacement policy that uses data dependency information to optimize cache management in data-parallel systems.

Findings

LRC improves cache hit ratio over LRU.

LRC speeds up applications by 60% in Spark.

LRC effectively utilizes DAG information for cache management.

Abstract

Memory caches are being aggressively used in today's data-parallel systems such as Spark, Tez, and Piccolo. However, prevalent systems employ rather simple cache management policies--notably the Least Recently Used (LRU) policy--that are oblivious to the application semantics of data dependency, expressed as a directed acyclic graph (DAG). Without this knowledge, memory caching can at best be performed by "guessing" the future data access patterns based on historical information (e.g., the access recency and/or frequency), which frequently results in inefficient, erroneous caching with low hit ratio and a long response time. In this paper, we propose a novel cache replacement policy, Least Reference Count (LRC), which exploits the application-specific DAG information to optimize the cache management. LRC evicts the cached data blocks whose reference count is the smallest. The reference count is defined, for each data block, as the number of dependent child blocks that have not been computed yet. We demonstrate the efficacy of LRC through both empirical analysis and cluster deployments against popular benchmarking workloads. Our Spark implementation shows that, compared with LRU, LRC speeds up typical applications by 60%.